LAUSR

This paper proposes a knowledge-based intelligent decision support system for operational risk management of global supply chains (DSSRMG), a full-phase system not yet treated in the literature. DSSRMG predicts the supply chain performance using the enhanced artificial neural network combined with particle swarm optimisation, infers the core risk source using a method based on principle component analysis, and evaluates risk mitigation alternatives using the digraph-matrix approach combined with principle component analysis. A methodology using an adaptive-network-based fuzzy inference system is suggested to construct the knowledge base for mitigation alternatives. An industrial example is used to illustrate the performance of DSSRMG. Computational experiments show that the techniques used for DSSRMG are excellent. Especially, the algorithm for selecting the useful operation indicators improves the performance prediction accuracy by 7.1% on average. DSSRMG provides supply chain managers with a practical tool to accurately predict and effectively control the operational risk. [Received: 9 March 2017; Revised: 22 July 2017; Accepted: 2 October 2017]